Mutations that play a role in the pathogenesis of human malignancies often occur in tumor suppressor genes (TSGs), which normally function to keep cell growth under strict control. Our team and collaborators have identified and cloned the transcription factor CTCF containing an 11 Zn finger (ZF) domain conserved from Drosophila to frogs to birds to mice to men. Different CTCF-target sites (CTSs), recognized by different combinations of CTCF ZFs, perform distinct regulatory functions. Depending on the context, different CTSs play distinct roles in transcriptional regulation by CTCF including promoter repression, activation, and creation of the thyroid hormone-responsive silencers. These studies resulted in more then 30 publications, and the "CTCF" Patent issued in 1999.Vertebrate chromatin insulators including the boundaries of the globin gene locus have been recently pinpointed by the Felsenfeld lab (NIDDKD, NIH) to be the different CTSs that are necessary and sufficient for CTCF-driven enhancer-blocking activity in a model system. Furthermore, in collaboration with the Ohlsson lab (Uppsala Univ., Sweden) we found and published that compared with the globin insulator CTSs, different CpG-containing subset of CTSs with an enhancer-blocking functions bind CTCF in the imprinting control region of the Igf2/H19 locus in vivo in a methylation-sensitive manner. Implications of this finding for gene imprinting and tumorigenesis associated with IGF2 activation will are very far-reaching and may have a significant impact on the whole field of Molecular Biology and Genetics of cancer as discussed in the reviews on CTCF just published in "Nature", "Science" and "Current Biology" journals. Among genes regulated through CTCF are important cell growth regulators including MYC, PIM-1, POLO-like kinase, and p19ARF. In the latter promoter, that is known to become frequently hypermethylated im tumors, CTCF binding is CpG-methylation dependent. In the MYC locus, CTCF appears to play a dual role by acting as a 5-prime chromatin boundary at the constitutive nuclease-hypersensitive site distant to the promoter region, and as a repressor immediately downstream of each of three MYC promoters. These findings may have an important impact on our understanding of molecular mechanisms of normal, and frequently dysregulated in cancer, MYC and/or p19ARF expression. Our characterization of functionally important CTCF targets in these genes suggested that acquired mutations in CTCF might be involved in cancer development. Supporting this, we mapped CTCF within a narrow cancer-associated "hot spot" on chromosome 16q22.1. A variety of cancers including some breast, prostate and Wilms' tumors display deletions at this locus accompanied by loss of imprinting, and/or deregulated MYC expression, and p19ARF aberrant methylation. The localization of the CTCF gene at 16q22.1 within the smallest region of overlap for losses of heterozygosity (LOH) often observed in breast, prostate, Wilms' and several other tumors and deregulation oCTCF-regulated target genes, such as Igf2 and MYC, frequently noticed in these tumors suggested an involvement of CTCF in neoplasia, i.e. that CTCF=TSG. This year we characterized several distinct somatic mutations of CTCF identified in breast, prostate and Wilms tumor cases with 16q22 LOH analyzed at exons coding for the CTCF 11-ZF-domain (one third of the entire protein). The mutations occurred within either of two ZFs and resulted in substitutions of amino acids at position critical for ZF formation or DNA base recognition. Each mutation abrogated or greatly diminished CTCF binding to different target sites in the genes governing cell growth (Igf2 insulator, and promoters of MYC, PIM-1, Polo-like kinase, and p19ARF) but did not change interaction of CTCF with targets found in growth-control-unrelated betta-globin insulator, lysozyme silencer or APP promoter. Our results also show that CTCF normally recognizes different sites by the combinatorial use of ZFs with the same individual ZFs sometimes being required for binding to one site but not another. This finding is unprecedented for multi-ZF proteins. Thus, unlike tumor-related mutations in other TSGs that lead to loss of function, mutations in CTCF are selectively dysfunctional, permitting wild-type binding to some sites while completely abrogating recognition of others. Thus, we obtained direct evidence that CTCF=TSG. CTCF mutations that shift the spectrum of binding specificities may thus represent a novel mechanism for tumor cell escape from growth control. This finding will have a predictable impact on molecular diagnostics and therapy of cancers associated with CTCF.